The current proposal seeks to perform therapeutic trials in subjects with AIDS-related lymphoma, with performance of correlative laboratory studies in order to determine various biologic factors involved in differing clinical responses. Using our previously existing strong multi- disciplinary collaborative network of clinicians, basic scientists and biostatisticians, and our large patient base, of which 40% have been of minority background, we hope to use data from results obtained to determine alternative new therapies for such patients. The following specific hypotheses will be addressed, by means of a prospective correlative therapeutic trial (Appendix #1) which employs m-BACOD chemotherapy with dideoxycytidine (ddC) and Granulocyte-CSF (r-met Hu G-CSF) support: 1) Effective chemotherapy together with effective anti-retroviral intervention will be required to achieve long-term survival in patients with AIDS-lymphoma. 2) The importance of decreasing HIV viral burden in this setting relates to potential improvement in survival by (a) decreasing complicating opportunistic infections and (b) decreasing the chance of lymphoma relapse, by decreasing the cascade of growth factors and cytokines produced by macrophages and lymphocytes in response to on-going HIV infection. The ability of ddC to reduce HIV viral burden will be assessed by serial determination of p24 antigenemia; quantitative cell associated viral load (% infected cells by culture of PBLs); quantitative plasma viremia (cell free virus burden by culture); and quantitative PCR, performed at baseline and serially. The correlation between HIV viral burden and cytokine expression will be assessed by means of quantifying the levels of TNF; IL- 1; and IL-6 in serum (as correlates of HIV proliferation); and IL-6; IL-4; IL-10 and B cell growth factor (BCGF-12kD) in tissues (as correlates of B cell proliferation), at baseline and serially. The significance of these biologic parameters will be assessed by means of correlation between clinical response, HIV viral burden and growth factor/cytokine expression. If these hypotheses are correct, future therapy could be focused on development and testing of various growth factor inhibitors.